4. SDO/AIA response to quiet Sun, active region and flare plasma

Author: Brendan O’Dwyer is a PhD student at the DAMTP, University of Cambridge.

<< previous nuggetnext nugget >>


The Atmospheric Imaging Assembly (AIA) on the Solar Dynamics Observatory (SDO) is a set of four telescopes designed to image the solar atmosphere in a variety of extreme ultraviolet (EUV), ultraviolet and visible-light wavelength bands. The instrument takes full-disk images, with high spatial resolution (~0.6 arcsec pixels) and with a cadence of 12 seconds. It is designed to observe solar plasma over a broad range of temperatures corresponding to the photosphere through to the corona. An example of an SDO/AIA image observed with the 193 A filter is shown in Figure 1. Each of the seven EUV filters is designed so that particularly strong spectral lines, characteristic of plasma at a particular temperature or range of temperatures, are transmitted.

Figure 1: An image of the Sun obtained with the SDO/AIA 193 A filter.

But the business of associating wavelength bands with temperatures is not a trivial one, since the wavelength range transmitted by each filter can contain several spectral lines produced by plasma at a range of temperatures. In this nugget we set out to check whether the plasma observed through each of the EUV channels has the temperature that we expect [1].

Of the seven EUV channels, six are expected to be dominated by emission lines from highly ionised iron: at 94 A (17-times ionised iron, or Fe XVIII), 131 A (Fe VIII, Fe XX, Fe XXIII), 171 A (Fe IX), 193 A (Fe XII, Fe XXIV), 211 A (Fe XIV), 335 A (Fe XVI), with the remaining channel 304 A dominated by He II. Using the CHIANTI spectral synthesis software, we have examined the contribution of spectral lines, and also continuum emission, to each of the EUV channels in order to determine what the dominant contribution to each channel is in different regions of the solar atmosphere.

Synthetic Spectra

Synthetic spectra are calculated using CHIANTI, taking as input the expected emission from a coronal hole, from quiet Sun, active regions and flares, in the form of differential emission measure curves. DEM curves give a measure of the amount of plasma as a function of temperature. The known spectral response of each AIA EUV channel – i.e. the filter and telescope transmission and the detector sensitivity as a function of wavelength – is then used to calculate the overall output produced by the channel (i.e. the synthetic spectra are convolved with the spectral response.) The results of this process are displayed in Figures 2-5, which show the contributions to four of the filters, represented by the intensity of the spectral lines.

Figure 2: Flare (blue) and QS (red) synthetic spectra for the 94 A channel. The spectral response is overplotted as a dashed line. The peak itensities of stronger lines are indicated. The weaker spectrum has been scaled by a factor indicated in the figure.
Figure 3: Same as Fig. 2 for the 131 A channel.
Figure 4: Same as Fig. 2 for the 211 A channel. The AR synthetic spectrum (green) is also included.
Figure 5: Same as Fig. 2 for the 335 A channel, except that the AR synthetic spectrum (green) is included instead of the flare spectrum.


In general our findings are in agreement with expectations, in that the lines which we expect to dominate in each filter do indeed provide strong contributions. However, in flares, the 131 A channel is predicted to be dominated by the Fe XXI 128.75 A line, and not by the Fe XX 132.84 A or Fe XXIII 132.91 A lines. In the quiet Sun, the dominant contribution to the 94 A channel is predicted to come from a line of Fe X at 94.01 A, and not from the much hotter Fe XVIII (although Fe XVIII does dominate in flares.)

Also in some instances significant contributions come not from lines at all, but from the continuum bremsstrahlung emission of hot plasma. This is true for example in the 131 A channel in active regions, and the 211 A channel in flaring regions.

So we see that care must be taken when interpreting SDO AIA images in terms of plasma temperature. In all channels, plasma at a range of temperatures contributes to the signal. Careful forward modeling and analysis techniques to determine the differential emission measure are really required. But even then one must be cautious, since much analysis – including the predictions presented here – are based on the properties of a plasma which is in local thermodynamic and ionisation equilibrium, and it is not clear that this will be the case, particularly in rapidly heating plasmas. However, used with care, the SDO AIA data have already, and will continue to enrich our view of the multi-thermal and dynamic corona.


[1] O’Dwyer, B., Del Zanna, G., Mason, H. E., Weber, M. A., Tripathi, D., 2010, A&A in press